Converting addresses to coordinates

# Converting addresses to coordinates

---

layout: true
  
<div class="dk-footer">
<span>
<a href="https://rfortherestofus.com/" target="_blank">R for the Rest of Us
</a>
</span>
</div>

---

class: center, middle, dk-section-title
background-image:url("images/street-sign.jpeg")
background-size: 100%

# Converting addresses to coordinates

???

---

## Addresses are messy

Street addresses are extremely messy and very *human* datasets.

- A building

- A floor in a building

- A collection of buildings

- A city/town/province or other region
]

Additionally:

- Addresses are often incomplete

- Addresses vary significantly internationally

]

???

---

## Geocoding: Forwards & Reverse

Geocoding is the process of converting between addresses and geographic coordinates.

- **Forward geocoding** *attempts* to converts an address into coordinates.

- **Reverse geocoding** *attempts* converts coordinates into an address.

???

---

## Geocoding packages

There are **several** packages that provide geocoding services - because there are multiple **commercial** geocoding services.

???

---

## `{ggmaps}`

Before June 2018 the most obvious and best package for geocoding was `{ggmaps}`:

- It used Google's industry leading Google Maps API

- It was free and easy to use

???

---

## `{ggmaps}`

Before June 2018 the most obvious and best package for geocoding was `{ggmaps}`:

- It used Google's industry leading Google Maps API

- It was free and easy to use

Nowadays you **must** provide Google with your **credit card details** to use this service.

- Currently you are provided $200 free credits per month

- Refer to Google's pricing pages for up to date details: [cloud.google.com/maps-platform/pricing](https://cloud.google.com/maps-platform/pricing)

???

---

## `{tidygeocoder}`

I strongly recommend using `{tidygeocoder}` for all your forward geocoding needs.

The package rolls together a total of **four** different geocoding services - but we only care about two of them.

???

---

## `{tidygeocoder}`

<div id="htmlwidget-41856a4abe521e595a57" style="width:100%;height:320px;" class="datatables html-widget"></div>
<script type="application/json" data-for="htmlwidget-41856a4abe521e595a57">{"x":{"filter":"none","vertical":false,"data":[["<img src=images/locationiq-logo.png width='400px'/>","<img src=images/openstreetmap-logo.png width='400px'/>"],["❌","❌"],["The free tier allows 5,000 requests/day and can be used commercially.","OpenStreetMap provide an always free geocoding service called <b>Nominatim<\/b>.<br>It's <b>not<\/b> designed for commercial usage, see the <a href='https://operations.osmfoundation.org/policies/nominatim/'>Usage Policy<\/a> for more details."]],"container":"<table class=\"display\">\n  <thead>\n    <tr>\n      <th>Geocoding service<\/th>\n      <th>Requires billing details?<\/th>\n      <th>Details<\/th>\n    <\/tr>\n  <\/thead>\n<\/table>","options":{"dom":"t","columnDefs":[{"className":"dt-center","targets":1},{"width":"200px","targets":1}],"order":[],"autoWidth":false,"orderClasses":false}},"evals":[],"jsHooks":[]}</script>

???

---

## My Turn

I'll import and geocode the addresses in the UK Addresses tab of our Excel file.

???

---

# (RSTUDIO CODING SLIDE)

???

---

## Environmental Variables

One of the most flexible ways to **permanently** store API keys on your machine.

The file is most easily manipulated with the `{usethis}` package:

```r
usethis::edit_r_environ()
```

???

---

## .Renviron

Variables are stored in the file as key-value pairs:

> LOCATIONIQ_API_KEY=KEY
>
> OTHER_API_KEY=KEY
>
> &zwnj;

The file **must** end with an empty line <span style="color:red">or it won't work</span>.

???

---

# (RSTUDIO CODING SLIDE)

???

---

## Your Turn

Convert the addresses in the "International Addresses" worksheet into coordinates and visualise them.

1. Import the "International Addresses" worksheet

1. Re-use the `mutate(across(...))` code to remove NA values

1. Combine together `address_line_1`, `address_line_2`, `address_line_3` into `full_address`

1. Use `geocode()` to convert the addresses into coordinates

1. Visualise the locations with `{mapview}`